Aims Potential advantages of real-time magnetic resonance imaging (MRI)-guided electrophysiology (MR-EP) include contemporaneous three-dimensional substrate assessment at the time of intervention, improved procedural guidance, and ablation lesion assessment. We evaluated a novel real-time MR-EP system to perform endocardial voltage mapping and assessment of delayed conduction in a porcine ischaemia-reperfusion model. Methods and results Sites of low voltage and slow conduction identified using the system were registered and compared to regions of late gadolinium enhancement (LGE) on MRI. The Sorensen-Dice similarity coefficient (DSC) between LGE scar maps and voltage maps was computed on a nodal basis. A total of 445 electrograms were recorded in sinus rhythm (range: 30-186) using the MR-EP system including 138 electrograms from LGE regions. Pacing captured at 103 sites; 47 (45.6%) sites had a stimulus-to-QRS (S- QRS) delay of >= 40ms. Using conventional (0.5-1.5mV) bipolar voltage thresholds, the sensitivity and specificity of voltage mapping using the MR-EP system to identify MR-derived LGE was 57% and 96%, respectively. Voltage mapping had a better predictive ability in detecting LGE compared to S-QRS measurements using this system (area under curve: 0.907 vs. 0.840). Using an electrical threshold of 1.5mV to define abnormal myocardium, the total DSC, scar DSC, and normal myocardium DSC between voltage maps and LGE scar maps was 79.0 +/- 6.0%, 35.0 +/- 10.1%, and 90.4 +/- 8.6%, respectively. Conclusion Low-voltage zones and regions of delayed conduction determined using a real-time MR-EP system are moderately associated with LGE areas identified on MRI.

In X-ray fluoroscopy, static overlays are used to visualize soft tissue. We propose a system for cardiac and respiratory motion compensation of these overlays. It consists of a 3-D motion model created from real-time magnetic resonance (MR) imaging. Multiple sagittal slices are acquired and retrospectively stacked to consistent 3-D volumes. Slice stacking considers cardiac information derived from the ECG and respiratory information extracted from the images. Additionally, temporal smoothness of the stacking is enhanced. Motion is estimated from the MR volumes using deformable 3-D/3-D registration. The motion model itself is a linear direct correspondence model using the same surrogate signals as slice stacking. In X-ray fluoroscopy, only the surrogate signals need to be extracted to apply the motion model and animate the overlay in real time. For evaluation, points are manually annotated in oblique MR slices and in contrast-enhanced X-ray images. The 2-D Euclidean distance of these points is reduced from 3.85 to 2.75 mm in MR and from 3.0 to 1.8 mm in X-ray compared with the static baseline. Furthermore, the motion-compensated overlays are shown qualitatively as images and videos.

Segmentation is one of the most important parts of medical image processing. Manual segmentation is very cumbersome and time-consuming. Fully automatic segmentation approaches require a large amount of labeled training data and may fail in difficult cases. In this paper, we propose a new method for 2-D segmentation and 3-D interpolation. The Smart Brush functionality quickly segments the ROI in a few 2-D slices. Given these annotated slices, our adapted formulation of Hermite Radial Basis Functions reconstructs the 3-D surface. Effective interactions with less number of equations accelerate the performance and therefore, a real-time and an intuitive, interactive segmentation can be supported effectively. The proposed method was evaluated on 12 clinical 3-D MRI data sets from individual patients and were compared to gold standard annotations of the left ventricle from a clinical expert. The 2-D Smart Brush resulted in an average Dice coefficient of 0.88± 0.09 for individual slices. For the 3-D interpolation using Hermite Radial Basis Functions an average Dice coefficient of 0.94± 0.02 was achieved.

Aims Magnetic resonance imaging (MRI) is the gold standard for defining myocardial substrate in 3D and can be used to guide ventricular tachycardia ablation. We describe the feasibility of using a prototype magnetic resonance-guided electrophysiology (MR-EP) system in a pre-clinical model to perform real-time MRI-guided epicardial mapping, ablation, and lesion imaging with active catheter tracking. Methods and results Experiments were performed in vivo in pigs (n = 6) using an MR-EP guidance system research prototype (Siemens Healthcare) with an irrigated ablation catheter (Vision-MR, lmricor) and a dedicated electrophysiology recording system (Advantage-MR, Imricor). Following epicardial access, local activation and voltage maps were acquired, and targeted radiofrequency (RF) ablation lesions were delivered. Ablation lesions were visualized in real time during RF delivery using MR-thermometry and dosimetry. Hyper-acute and acute assessment of ablation lesions was also performed using native T1 mapping and late-gadolinium enhancement (LGE), respectively. High-quality epicardial bipolar electrograms were recorded with a signal-to-noise ratio of greater than 10:1 for a signal of 1.5 mV. During epicardial ablation, localized temperature elevation could be visualized with a maximum temperature rise of 35 degrees C within 2 mm of the catheter tip relative to remote myocardium. Decreased native T1 times were observed (882 +/- 107 ms) in the lesion core 3-5 min after lesion delivery and relative location of lesions matched welt to LGE. There was a good correlation between ablation lesion site on the iCMR platform and autopsy. Conclusion The MR-EP system was able to successfully acquire epicardial voltage and activation maps in swine, deliver, and visualize ablation lesions, demonstrating feasibility for intraprocedural guidance and real-time assessment of ablation injury.

Layered motion estimation (LME) in X-ray fluoroscopy is a challenging, ill-posed and non-convex problem due to transparency effects and the way the image is defined. Minimizing an energy formulation of layered motion estimation is computationally expensive. For clinical usability of this approach, we propose to use primal-dual optimization parallelized using a graphical processing unit (GPU) to reduce the overall run-time of this algorithm. Experimentally this method is able to substantially reduce target registration error by 70% on manually annotated landmarks on five distinct image sequences compared to the static baseline, similar to prior work on this domain. However, the overall runtime of our method on a conventional GPU is less than 3.3 seconds compared to several minutes for the state of the art. Considering typical framerates of X-ray fluoroscopy devices, this runtime makes the application of layered motion estimation feasible for many clinical workflows.

Respiratory signals are required for image gating and motion compensation in minimally invasive interventions. In X-ray fluoroscopy, extraction of a respiratory signal can be challenging due to characteristics of interventional imaging, in particular injection of contrast agent and automatic exposure control. We present a novel method for respiratory signal extraction based on dimensionality reduction that can tolerate these events. Images are divided into patches of multiple sizes. Low-dimensional embeddings are generated for each patch using illumination-invariant kernel PCA. Patches with respiratory information are selected automatically by agglomerative clustering. The signals from this respiratory cluster are combined robustly to a single respiratory signal. In the experiments, we evaluate our method on a variety of scenarios. If the diaphragm is visible, we track its superior-inferior motion as ground truth. Our method has a correlation coefficient of more than 91% with the ground truth irrespective of whether or not contrast agent injection or automatic exposure control occur. Additionally, we show that very similar signals are estimated from biplane sequences and from sequences without visible diaphragm. Since all these cases are handled automatically, the method is robust enough to be considered for use in a clinical setting.

Motion estimation in X-ray images is a challenging task due to transparently overlapping structures from different depths. We propose to separate an X-ray sequence into a static and a dynamic layer to facilitate motion estimation. The method exploits the idea to use the minimum intensity over time and a spatial smoothness prior for both layers. For numerical optimization, we propose a conditional Markov random field. In experiments on synthetic data, we achieve a root mean squared intensity difference of 36.7±8.4 to the ground truth static layer. In addition, we show qualitative results that demonstrate an improved layer separation compared to state-of-the-art algorithms.

Fluoroscopic images are characterized by a transparent projection of 3-D structures from all depths to 2-D. Differently moving structures, for example due to breathing and heartbeat, can be described approximately using independently moving 2-D layers. Separating the fluoroscopic images into the motion layers is desirable to facilitate interpretation and diagnosis. Given the motion of each layer, it is state of the art to compute the layer separation by minimizing a least-squares objective function. However, due to high noise levels and inaccurate motion estimates, the results are not satisfactory in X-ray images. In this work, we propose a probabilistic model for motion layer separation. In this model, we analyze various data terms and regularization terms theoretically and experimentally. We show that a robust penalty function is required in the data term to deal with noise and shortcomings of the image formation model. For the regularization term, we propose to enforce smoothness of the layers using bilateral total variation. On synthetic data, the mean squared error between the estimated layers and the ground truth is improved by 18 % compared to the state of the art. In addition, we show qualitative improvements on real X-ray data.

Dense motion estimation in X-ray fluoroscopy is challenging due to low soft-tissue contrast and the transparent projection of 3-D information to 2-D. Motion layers have been introduced as an intermediate representation, but so far failed to generate plausible motions because their estimation is ill-posed. To attain plausible motions, we include prior information for each motion layer in the form of a surrogate signal. In particular, we extract a respiratory signal from the images using manifold learning and use it to define a surrogate-driven motion model. The model is incorporated into an energy minimization framework with smoothness priors to enable motion estimation. Experimentally, our method estimates 48% of the 2-D motion field on XCAT phantom data. On real X-ray sequences, the target registration error of manually annotated landmarks is reduced by 52%. In addition, we qualitatively show that a meaningful separation into motion layers is achieved.

In X-ray fluoroscopy-guided minimally invasive interventions, overlays of pre-procedurally acquired image data can be used to visualize soft-tissue. In the thoracic and abdominal regions, static overlays are inconsistent to the live X-ray images due to respiratory motion of the patient. This error can be reduced by dynamically adapting the overlay to the respiration. A first step in this direction is the real-time extraction of the respiratory state from the live X-ray images. The respiratory state can drive a motion model to compensate the breathing motion. We present a method to extract respiratory signals from X-ray sequences in real-time. Respiratory signal extraction is viewed as a dimensionality reduction problem, which is performed for each X-ray image using incremental Isomap. The method has a correlation of 0.97 ± 0.02 with internal breathing motion and an average runtime of 42 ms per image. The method is accurate, robust, and can be used in a wide range of clinical applications and fields of view.

Rigid registration is a key requirement to overlay anatomical structures from pre-operative magnetic resonance (MR) images onto live X-ray fluoroscopic images. Clinical applications are for instance radio frequency catheter ablation and transcatheter endomyocardial injection. We present a method for fully automatic marker-based detection and registration. The algorithm locates the markers in the MR volumetric data set automatically by combining homomorphic unsharp masking, adaptive thresholding, region growing, and shape analysis. In the X-ray images, the markers are localized in several 2-D images and then their 3-D positions are reconstructed. The 2-D X-ray localization is performed using adaptive thresholding and template matching. The 3-D X-ray marker positions are established using back projection. Rigid registration between the two 3-D point sets is performed by an iterative closest point algorithm. The 3-D registration error using a flat panel detector X-ray system is on average 1.5 ± 1.8 mm. The novelty of our approach is the incorporation of bundle adjustment to reduce the 2-D overlay errors. Bundle adjustment reduces the 2-D X-ray marker localization error from 0.6 ± 0.8 mm to 0.4 ± 0.5 mm. The target registration error of our proposed method evaluated on eight datasets is on average 1.4±3.5 mm. These targets were not used during registration. Accordingly, anatomical structures from MR are positioned more accurately.